Online condition monitoring wind turbine by the approach of local empirical mode decomposition Authors

This research introduces the local and online smoothing sifting process for the EMD approach, as a substitute for the traditional sifting process. In this method, the local mean value of the signal at each point is extracted by applying smoothing filters to its adjacent data points, within a variable span sliding window. This approach is direct, local and online; hence, it can improve the EMD performance, and overcome many drawbacks of the classical EMD algorithm. The effectiveness of the proposed approach has been validated by using experimental data which is discussed in the paper. It is these same experiments show that local EMD can potentially be a powerful tool for conducting online wind turbine condition monitoring.